Solutions Architect @ NVIDIA · HPC · AI Infrastructure · Energy Efficiency
Santa Clara, CA

I design and optimize large-scale AI systems at NVIDIA, focusing on:

• 🏗️ HPC + AI infrastructure design (multi-node, scheduling, deployment)
• ⚡ LLM inference optimization (vLLM · TensorRT-LLM · DeepSpeed)
• 🔥 Performance-per-watt benchmarking across GPU clusters

How fast can we run — and at what energy cost?

• NeurIPS 2026 · Accepted
  The 1/W Law: Context Length is the Dominant Energy Lever in LLM Inference Fleets
  Proceedings forthcoming.

• ICCAD 2026 · Accepted
  MicroEvo: Knowledge-Guided LLM Sampling for Efficient Microarchitecture Design Space Exploration
  Preprint · Code
  Proceedings forthcoming.

• AAAI 2026
  TokenPowerBench: Benchmarking the Power Consumption of LLM Inference
  Paper · Code

• AAAI 2026
  FIXME: Towards End-to-End Benchmarking of LLM-Aided Design Verification
  Paper

More publications · Google Scholar

• Multi-node GPU benchmarking across:
  • vLLM · TensorRT-LLM · DeepSpeed · HuggingFace
• Throughput · Latency · Memory · Power efficiency

• Performance-per-watt analysis across workloads
• Sustainable AI system design
• Trade-offs: performance vs energy cost

• Slurm-based HPC scheduling
• Kubernetes-based AI deployment
• Hybrid cluster orchestration
• Multi-node GPU scaling

AI / ML

GPU / Systems

Infrastructure

Languages

• LLM inference engine benchmarking
• GPU power and energy-efficiency profiling
• Multi-node AI cluster performance analysis
• HPC scheduling and AI infrastructure design
• Sustainable AI system optimization

Website · LinkedIn · Google Scholar · ORCID